Interpretation of Nuclear Magnetic Resonance Measurements in Formations with Complex Pore Structure

Nuclear Magnetic Resonance (NMR) measurement as an unconventional well logging method, is among the most accurate approaches to estimate formation porosity. Petrophysicists use NMR to evaluate petrophysical properties such as pore size distribution, permeability, and fluid saturation, of various formations. However, the NMR responses in complex formations, such as fractured carbonate and organicrich mudrocks, have not been thoroughly investigated. NMR pore-scale simulations using a random-walk algorithm enable us to model the NMR relaxometry in porous rock samples, and to improve interpretation of NMR relaxometry in complex formations. Based on pore-scale simulations and theoretical analysis of NMR relaxometry, this research estimated petrophysical properties that have been challenging when using conventional NMR interpretation, including micro-fracture volumetric concentration, directional pore connectivity and directional permeability, and the impact of wettability alteration. This dissertation (a) quantified the impacts of micro-fractures on NMR relaxation times by pore-scale simulations and developed an analytical model for fracture-pore diffusional coupling in multiple-pore-type systems (i.e. composed of intra-/inter-granular pores and micro-fractures); (b) investigated the viability of using the NMR analytical model to estimate the volumetric concentrations and apertures of micro-fractures in fractured formations; (c) developed an innovative NMR-based directional permeability model to estimate anisotropic permeability of rock samples with complex pore structure; (d) investigated the impacts of fracturepore diffusional coupling on NMR permeability assessment and evaluated reliability of NMR permeability models in fractured formations; and (e) developed a two-phase NMR pore-scale simulation method to model the NMR responses in organicrich mudrocks, as well as investigated the impact of wettability alteration on NMR relaxometry in organic-rich mudrocks. The methods used in this research include pore-scale image processing, single-phase and two-phase NMR simulations in porous media, fluid flow simulations in porous media, and theoretical analysis of NMR relaxation mechanisms in porous media. Results show that the introduced methods for interpretation of NMR relaxometry can enhance reservoir characterization in challenging reservoirs, including carbonates and organic-rich mudrocks